Brit.J.Psychia:. (:977),131,43:-2
Correspondence Letters for publication in the Correspondence columns should not ordinarily be more than 500 words and should be addressed to:
The Editor, British Journal HAMILTON
RATiNG
SCALE
of Psychiatry,
i7 Beigrave Square, London SWiX 8PG
FOR
sub-scales
DEPRESSION
(‘Bech' and
‘¿non-Beth') derived
from
it,
et al (Journal, July 1977,
all correlate to about the same degree with global ratings. The HRS is now a well-tried instrument. Like a
pp 49—52)comments upon a paper by Bech and his
good medicine, its staying power probably attests
co-workers in Denmark (I). The Danish paper sought to show that the Hamilton Rating Scalefor De
further work on improvement
pression
should cease. Both groups of workers have recog
DEAR
SIR,
The paper by Knesevich
(HRS)
is not a valid
measuring
to its effectiveness, but this should not imply that
instrument
for depressive illness throughout the full range of
and modification
nized this. However, there is a major point which the Danish workers appear to have overlooked. At the
severity, since scores failed to distinguish the more severe degree of illness. The paper by Knesevich et al,
time of its inception, Hamilton was concerned that his scale should reflect not only the severity of a
however, purports to refute the Danish findings. Unfortunately, both groups of workers have fallen into certain methodological errors, and it is im
depressive illness but the total symptom profile. By
portant
and his co-workers the scale may well be easier to
that these
should
be recognized
reducing
lest future
workers in this field repeat them. Both groups have
use and so commend itself to some researchers and clinicians but thisdoes not indicatethat the ab breviated scale is always the more appropriate
used repeated measures from the same patients, and this really should not be allowed, even though ratings were included at different stages of the illness.
instrument.
Both groups purport to have tested the HRS through out the full range of severity of depressive illness, but both have failed to include any patients with a severity
greater
than
the ninth
grade
to note that inspection
of the Figure in
score at the ninth grade is
made
both
the global
ratings
and
‘¿pure' depressive
with anxiety
symptoms
pain whilst ignoring limitation of movement and joint swelling. There may well be occasions on which an abbreviated scale may be more appropriate but this may not always be the case. (An occasion for the use of partial Hamilton scores is stated in one of our own papers
(2).)
There can be no doubt thatthe most accurate
measures of the severity of depressive (and anxiety) states are based upon composite scores, in which observer ratings such as those of Hamilton achieve an
have avoided the pitfall. It concerns the fact that the workers
illness,
in Hamilton's scale. If in future the severity of the illness is to be determined by a measure of just one group of symptoms it will be equivalent to a measure of the severity of rheumatoid arthritis based solely on
on an eleven
about 21 @2 i.e. lower scores at the more severe degree as judged on the global rating. The final and most severe criticism applies to the American paper, while the Danish workers appear to same
a depressive
is intermingled
and somatic symptoms and this fact is recognized
the paper by Knesevich et al shows that the mean HRS score at the eighth grade of severity is about 24@7 whereas the mean
In
symptomology
point scale. It is perhaps understandable that they have failed to do so, for severe retardation or agitation causes difficulty in making ratings. rn this connection it is important
the scale to six items as suggested by Bech
equal
the
weight
with the patients
own view of the
Hamilton ratings (if this is not so, it is not stated in
severity of his state, the self-rating, and a third
the paper). HRS scores are subject to influence by the rater's general impression of how ill he considers
measure based upon an independent global assess ment of severity. Any two or even all of these measures when combined, will give a better overall assess
the patient to be and therefore work based upon validatingone measure against the other demands that the global and the Hamilton rating are made by
ment of severity than any one alone.
different workers and that they do nothing to betray
Department of P@ychiahy, The Univers4y of Leeds,
their rating to each other until the final analysis. This error may well account for the finding in the American
paper
that
the
total
HRS
and
the
15 Hyde
Terrace,
Leeds,
LS2 9LT
R. P. SNAITH
two 43!
CORRESPONDENCE
432 REFERENCES
are
:. BECH, P., GRAM, L. F., DEIN, E., JACOBSEN, 0., VITGER, J. and BOLWIG, T. G. (:975) Quantitative
Rating of Depressive navwa, 51, :61—70.
2.
States. Ada Psychiatrica Scan
differentiated
simply
by
the
presence
of
‘¿(,) those with a preponderance
of delusions of
contrition over delusions of grandeur or of perse SNAJTH, R. P., Constantopoulos, A. A.,JAIwnen, M. Y., cution may, in the absence of delusions of disinte McGUFFIN, P. (:g77) A Clinical Scale for the gration be regarded as virtually synonymous with Self-Assessment of Irritability. British Jownal @f Psychotic Depressives; (2) dCs (or Psychotic Depres Psychiatry. (In Press). sives) almost invariably suffer from neurotic symptoms
THE IMPACT OF AN ABNORMAL CHILD
UPON THE PARENTS
Dn@sitSm, In the April 1977 issue of this Journal (130, pp 405— :o), Dr Gath comments of the marital
relationship'
of the parents
May
Dr Gath
of mongols,
absence
of
neurotic
symptoms
(phobic,
of Classes3, 2, and :, sDs
are only members of Class i. There are, therefore, differentiae at two class levels. Bagshaw
nosis of psychotic (dC) or neurotic (sD) depression'. A further twenty-two cases (29%) fell into one of the
descriptions
neurotic symptoms
but
careful
Hirsh and Leff (i@7@) later revealed
into
words, dCs are members
diagnosed
ink-blots,
has fallen
the
the
of
that
in
same trap as did Singer and Wynne (1963) in their study of parents of schizophrenics? It will be recalled that the latter were found to give more pathological
@
I suggest
(with ruminations the commonest); (@) all dCs (or Psychotic Depressives) suffer from either a state of anxiety or of depression and usually both; (ii) a state of depression, as a diagnosLc, can only be identified
‘¿quality hysterical, or obsessional) and of delusions.' In other
upon the impaired
as evidence by their admissions of ‘¿severe tension, high hostility or marked lack of warmth' at interview.
studies
by
that the dif
found
case for separating neurotic
of parents
with prob
constructive
or destructive
to the marriage,
while
control parents related more moderate feelings. This may reflect the anxiety of the former to enlist help, rather than supporting Dr Gath's hypothesis.
A. C. CARR 77z Maudsie, Hospital,
% of her
‘¿psychiatrically
a DSSI diag
groups. This is precisely the area
of confusion we oommented upon when we made a
to talk at interview
on the part
that
depressed sample obtained
ference resulted simply from the increased willingness lem children. In Dr Gath's study, parents of mongols confessed more frequently to all the measures assessed, whether
@
not
delusions of contrition in one group. We most re cently concluded (Foulds and Bedford, :976a) that
As 92%
out the dysthysnic states from the
symptoms
(Foulds
and
Bedford,
:@i6b).
of her patientsfittedthe hierarchy model it
would seem plausible that the agreement in her study
between the clinical and DSSI ‘¿diagnoses' could be as high as 71 Finally, unpublished data front a pilot study of self-reinforcement
in
depression
has
been
made
available to me by Mr RJ Wycherley. 93% of his 40 in-patients have DSSI patterns which fit the hierarchy
model.Of the28 scoringsD i6 arealsodC
°%J;
of the i9 scoring dC i6 are also sD (8i54J. Again the
Denmark Hill, London SE5
relationship between these two sets might reasonably be described as inclusive and non-reflexive.
REFERENcFa
Wvroa@, L. C. and SINGER, M. T. (:963)Thought dis
order and family relations of schizopb.renics. Archives of GeneralPsychiatry,9, 191-206. Hmscis, S. R. and lm, J. P. (:975) Abnormalities in
ALAN B@D Warlinghans Park Hospital, Warlingham, Surrey CR3 9TR
Parents of Schizophrenics. Maudsley Monograph No. 22. London:
Oxford
University
Press.
THE HIERARCHICAL DEAR
REPERENcFS
MODEL
B@osIIAw,V. E. (:977) A replication study of Foulds' and Bedford's hierarchical model of depression.
OF
British Journal of Psychiatry, 131, 53-5; and 56-8.
DEPRESSION
Sm,
Fouus, G. A. & Banroiw, A. (:976a) Classification of depressive
It is comforting to have one's work replicated (Bagshaw, :977) and to see this twice in one Journal issue suggests something more than ‘¿sheer careless ness'. I wish to point out, however, that the author also
comes
close
inadvertantly
to
replicating
the
error of Kendell (1976). Within the hierarchical model of personal illness the ‘¿two types of depression'
illness:
a
re-evaluation.
Psychological
Medicine, 6, 15—9. —¿
(1976b)
The
relationship
depression and the neuroses. Psychiatry, zz8, s66—8. KENDELL,
it E. (1976)
The
a review of contemporary Psychiatry, 129, 15—28.
between
British
anxiety.
Journal
of
classification of depressions.:
confusion. British JournaLqf
Hamilton rating scale for depression.
R P Snaith BJP 1977, 131:431-432. Access the most recent version at DOI: 10.1192/bjp.131.4.431
References Reprints/ permissions You can respond to this article at Downloaded from
This article cites 0 articles, 0 of which you can access for free at: http://bjp.rcpsych.org/content/131/4/431.citation#BIBL To obtain reprints or permission to reproduce material from this paper, please write to [email protected] /letters/submit/bjprcpsych;131/4/431 http://bjp.rcpsych.org/ on May 29, 2016 Published by The Royal College of Psychiatrists
To subscribe to The British Journal of Psychiatry go to: http://bjp.rcpsych.org/site/subscriptions/
Correspondence Letters for publication in the Correspondence columns should not ordinarily be more than 500 words and should be addressed to:
The Editor, British Journal HAMILTON
RATiNG
SCALE
of Psychiatry,
i7 Beigrave Square, London SWiX 8PG
FOR
sub-scales
DEPRESSION
(‘Bech' and
‘¿non-Beth') derived
from
it,
et al (Journal, July 1977,
all correlate to about the same degree with global ratings. The HRS is now a well-tried instrument. Like a
pp 49—52)comments upon a paper by Bech and his
good medicine, its staying power probably attests
co-workers in Denmark (I). The Danish paper sought to show that the Hamilton Rating Scalefor De
further work on improvement
pression
should cease. Both groups of workers have recog
DEAR
SIR,
The paper by Knesevich
(HRS)
is not a valid
measuring
to its effectiveness, but this should not imply that
instrument
for depressive illness throughout the full range of
and modification
nized this. However, there is a major point which the Danish workers appear to have overlooked. At the
severity, since scores failed to distinguish the more severe degree of illness. The paper by Knesevich et al,
time of its inception, Hamilton was concerned that his scale should reflect not only the severity of a
however, purports to refute the Danish findings. Unfortunately, both groups of workers have fallen into certain methodological errors, and it is im
depressive illness but the total symptom profile. By
portant
and his co-workers the scale may well be easier to
that these
should
be recognized
reducing
lest future
workers in this field repeat them. Both groups have
use and so commend itself to some researchers and clinicians but thisdoes not indicatethat the ab breviated scale is always the more appropriate
used repeated measures from the same patients, and this really should not be allowed, even though ratings were included at different stages of the illness.
instrument.
Both groups purport to have tested the HRS through out the full range of severity of depressive illness, but both have failed to include any patients with a severity
greater
than
the ninth
grade
to note that inspection
of the Figure in
score at the ninth grade is
made
both
the global
ratings
and
‘¿pure' depressive
with anxiety
symptoms
pain whilst ignoring limitation of movement and joint swelling. There may well be occasions on which an abbreviated scale may be more appropriate but this may not always be the case. (An occasion for the use of partial Hamilton scores is stated in one of our own papers
(2).)
There can be no doubt thatthe most accurate
measures of the severity of depressive (and anxiety) states are based upon composite scores, in which observer ratings such as those of Hamilton achieve an
have avoided the pitfall. It concerns the fact that the workers
illness,
in Hamilton's scale. If in future the severity of the illness is to be determined by a measure of just one group of symptoms it will be equivalent to a measure of the severity of rheumatoid arthritis based solely on
on an eleven
about 21 @2 i.e. lower scores at the more severe degree as judged on the global rating. The final and most severe criticism applies to the American paper, while the Danish workers appear to same
a depressive
is intermingled
and somatic symptoms and this fact is recognized
the paper by Knesevich et al shows that the mean HRS score at the eighth grade of severity is about 24@7 whereas the mean
In
symptomology
point scale. It is perhaps understandable that they have failed to do so, for severe retardation or agitation causes difficulty in making ratings. rn this connection it is important
the scale to six items as suggested by Bech
equal
the
weight
with the patients
own view of the
Hamilton ratings (if this is not so, it is not stated in
severity of his state, the self-rating, and a third
the paper). HRS scores are subject to influence by the rater's general impression of how ill he considers
measure based upon an independent global assess ment of severity. Any two or even all of these measures when combined, will give a better overall assess
the patient to be and therefore work based upon validatingone measure against the other demands that the global and the Hamilton rating are made by
ment of severity than any one alone.
different workers and that they do nothing to betray
Department of P@ychiahy, The Univers4y of Leeds,
their rating to each other until the final analysis. This error may well account for the finding in the American
paper
that
the
total
HRS
and
the
15 Hyde
Terrace,
Leeds,
LS2 9LT
R. P. SNAITH
two 43!
CORRESPONDENCE
432 REFERENCES
are
:. BECH, P., GRAM, L. F., DEIN, E., JACOBSEN, 0., VITGER, J. and BOLWIG, T. G. (:975) Quantitative
Rating of Depressive navwa, 51, :61—70.
2.
States. Ada Psychiatrica Scan
differentiated
simply
by
the
presence
of
‘¿(,) those with a preponderance
of delusions of
contrition over delusions of grandeur or of perse SNAJTH, R. P., Constantopoulos, A. A.,JAIwnen, M. Y., cution may, in the absence of delusions of disinte McGUFFIN, P. (:g77) A Clinical Scale for the gration be regarded as virtually synonymous with Self-Assessment of Irritability. British Jownal @f Psychotic Depressives; (2) dCs (or Psychotic Depres Psychiatry. (In Press). sives) almost invariably suffer from neurotic symptoms
THE IMPACT OF AN ABNORMAL CHILD
UPON THE PARENTS
Dn@sitSm, In the April 1977 issue of this Journal (130, pp 405— :o), Dr Gath comments of the marital
relationship'
of the parents
May
Dr Gath
of mongols,
absence
of
neurotic
symptoms
(phobic,
of Classes3, 2, and :, sDs
are only members of Class i. There are, therefore, differentiae at two class levels. Bagshaw
nosis of psychotic (dC) or neurotic (sD) depression'. A further twenty-two cases (29%) fell into one of the
descriptions
neurotic symptoms
but
careful
Hirsh and Leff (i@7@) later revealed
into
words, dCs are members
diagnosed
ink-blots,
has fallen
the
the
of
that
in
same trap as did Singer and Wynne (1963) in their study of parents of schizophrenics? It will be recalled that the latter were found to give more pathological
@
I suggest
(with ruminations the commonest); (@) all dCs (or Psychotic Depressives) suffer from either a state of anxiety or of depression and usually both; (ii) a state of depression, as a diagnosLc, can only be identified
‘¿quality hysterical, or obsessional) and of delusions.' In other
upon the impaired
as evidence by their admissions of ‘¿severe tension, high hostility or marked lack of warmth' at interview.
studies
by
that the dif
found
case for separating neurotic
of parents
with prob
constructive
or destructive
to the marriage,
while
control parents related more moderate feelings. This may reflect the anxiety of the former to enlist help, rather than supporting Dr Gath's hypothesis.
A. C. CARR 77z Maudsie, Hospital,
% of her
‘¿psychiatrically
a DSSI diag
groups. This is precisely the area
of confusion we oommented upon when we made a
to talk at interview
on the part
that
depressed sample obtained
ference resulted simply from the increased willingness lem children. In Dr Gath's study, parents of mongols confessed more frequently to all the measures assessed, whether
@
not
delusions of contrition in one group. We most re cently concluded (Foulds and Bedford, :976a) that
As 92%
out the dysthysnic states from the
symptoms
(Foulds
and
Bedford,
:@i6b).
of her patientsfittedthe hierarchy model it
would seem plausible that the agreement in her study
between the clinical and DSSI ‘¿diagnoses' could be as high as 71 Finally, unpublished data front a pilot study of self-reinforcement
in
depression
has
been
made
available to me by Mr RJ Wycherley. 93% of his 40 in-patients have DSSI patterns which fit the hierarchy
model.Of the28 scoringsD i6 arealsodC
°%J;
of the i9 scoring dC i6 are also sD (8i54J. Again the
Denmark Hill, London SE5
relationship between these two sets might reasonably be described as inclusive and non-reflexive.
REFERENcFa
Wvroa@, L. C. and SINGER, M. T. (:963)Thought dis
order and family relations of schizopb.renics. Archives of GeneralPsychiatry,9, 191-206. Hmscis, S. R. and lm, J. P. (:975) Abnormalities in
ALAN B@D Warlinghans Park Hospital, Warlingham, Surrey CR3 9TR
Parents of Schizophrenics. Maudsley Monograph No. 22. London:
Oxford
University
Press.
THE HIERARCHICAL DEAR
REPERENcFS
MODEL
B@osIIAw,V. E. (:977) A replication study of Foulds' and Bedford's hierarchical model of depression.
OF
British Journal of Psychiatry, 131, 53-5; and 56-8.
DEPRESSION
Sm,
Fouus, G. A. & Banroiw, A. (:976a) Classification of depressive
It is comforting to have one's work replicated (Bagshaw, :977) and to see this twice in one Journal issue suggests something more than ‘¿sheer careless ness'. I wish to point out, however, that the author also
comes
close
inadvertantly
to
replicating
the
error of Kendell (1976). Within the hierarchical model of personal illness the ‘¿two types of depression'
illness:
a
re-evaluation.
Psychological
Medicine, 6, 15—9. —¿
(1976b)
The
relationship
depression and the neuroses. Psychiatry, zz8, s66—8. KENDELL,
it E. (1976)
The
a review of contemporary Psychiatry, 129, 15—28.
between
British
anxiety.
Journal
of
classification of depressions.:
confusion. British JournaLqf
Hamilton rating scale for depression.
R P Snaith BJP 1977, 131:431-432. Access the most recent version at DOI: 10.1192/bjp.131.4.431
References Reprints/ permissions You can respond to this article at Downloaded from
This article cites 0 articles, 0 of which you can access for free at: http://bjp.rcpsych.org/content/131/4/431.citation#BIBL To obtain reprints or permission to reproduce material from this paper, please write to [email protected] /letters/submit/bjprcpsych;131/4/431 http://bjp.rcpsych.org/ on May 29, 2016 Published by The Royal College of Psychiatrists
To subscribe to The British Journal of Psychiatry go to: http://bjp.rcpsych.org/site/subscriptions/
